MCP HeartbeatTHE MCP OBSERVATORY
Public workspace

Explore

Evidence for your next connection.

Independent measurements of remote MCP servers. Every observation has a timestamp.

Read-only probes

MCP server · Code & DevOps

Andes Protocol

Public procurement for AI agents: Chile tenders and Compra Ágil, fit, buyer and supplier history.

Checks passed
12 / 22

2 failed · 8 not measured

Median response
1130 ms

Median of successful probes over 7 days

30-day uptime
Not measured

Share of countable probes over 30 days, including daily rollups. Auth challenges are excluded.

Discovered tools
7

Tool definitions returned by this server

Independent monitoring

Health & checks

Status history

Responded

The server completed its last measured handshake. This is an observation, not a guarantee.

Last checked
Check summary by test group
Test areaPassedFailedUnmeasured
ConnectionCan a client connect and discover tools?
500
Reliability & speedDoes it respond consistently and quickly?
004
Access & authenticationWhat is required to use this server?
102
Tool qualityAre the tool definitions usable by an agent?
310
Publisher identityCan the publisher be traced to its source?
312

Unmeasured means there is not enough evidence to decide. How these checks work

Inspect individual checks 22 recorded

Core

  • Pass
    Reachable

    Latest probe returned HTTP 200 in 1130 ms.

  • Pass
    Valid handshake

    protocolVersion 2025-11-25, serverInfo.name andesprotocol.

  • Pass
    Lists tools

    tools/list returned 7 tools.

  • Pass
    Current protocol

    Negotiated 2025-11-25. The bar is 2025-06-18 or newer.

  • Pass
    HTTPS

    The endpoint URL is https.

Performance

  • Not measured
    Fast (p50)

    Median latency 1130 ms over 1 samples in 7 days. The bar is 1000 ms. Need at least 5 samples.

  • Not measured
    Consistent (p95)

    95th percentile 1130 ms over 1 samples in 7 days. The bar is 3000 ms. Need at least 20 samples.

  • Not measured
    Reliable (uptime)

    1 of 1 attempts returned a tool list in 30 days (100%). The bar is 99%. Need at least 30 samples.

  • Not measured
    Stable tool list

    Fewer than two successful probes in 30 days, so tool-list stability is not measured.

Access

  • Pass
    Open access

    tools/list succeeded with no credential, and the record declares none.

  • Not measured
    Standard auth

    Auth is not required, so the OAuth metadata check does not apply.

  • Not measured
    Free tier cited

    No publisher quote has been checked.

Tools

  • Pass
    Tools described

    Every tool description is at least 20 characters.

  • Pass
    Valid input schemas

    Every inputSchema is a JSON Schema object.

  • Fail
    Parameters documented

    0 of 35 parameters have a description (0%). The bar is 80%.

  • Pass
    Safety hints

    Every tool sets readOnlyHint or destructiveHint.

Publisher

  • Pass
    Verified namespace

    Listed in the Official MCP Registry.

  • Pass
    Website matches publisher

    The website host matches the reverse-domain namespace.

  • Fail
    Source repo linked

    No repository URL is on file.

  • Not measured
    Repo owner matches

    This namespace is not an io.github login, so the owner comparison does not apply.

  • Pass
    Recently updated

    Updated 2026-11-05T18:49:58.115066Z, within 180 days.

  • Not measured
    Open-source licence

    No GitHub repository to ask about a licence.

Start using Andes Protocol

Connect to this server

Choose your client, copy the configuration, then enable the server in your client’s MCP settings.

Run this command in your terminal, then restart Claude Code.

Claude Code

claude mcp add --transport http andes-protocol https://mcp.andesprotocol.com/mcp

Capabilities

Tools 7

assess_fit

Use this to decide whether a tender is worth your user's time. Pass tender_id plus company_profile (sector, capabilities, brands, certifications, in the user's words) and/or supplier_id (the user's tax id, e.g. Chilean RUT). Andes checks evidence it holds: the company's terms (with procurement synonyms) in the tender's title, items and description; the supplier's past awards in the same UNSPSC product category; the estimated amount against the supplier's usual award; and whether it already sold to this buyer. Returns fit (strong, possible, weak, insufficient_evidence), relevance_score 0-1, signals (each with status, strength, evidence and source), gaps (what would improve the judgment), the buyer's competition context (average bidders, usual winners) and agent_guidance. Without supplier_id the score is capped: text cannot prove capability. It does not read tender documents yet. Treat it as evidence; the decision is your user's. Access: needs an identified agent key (free, self-service at https://pass.andesprotocol.com).

Discovered
buyer_history

Use this to learn how a buyer behaves before bidding: what it awarded, to which suppliers, how many bidders its tenders draw (median_bidders, single_bidder_share) and how awards compare with the estimate (median_award_to_estimate). Pass tender_id from search_tenders, or buyer_name, or buyer_id. Built from the buyer's awarded tenders in Open Contracting (OCDS) data; period, sources and computed_at are in each match. An empty matches list comes with a note saying why. Access: needs an identified agent key (free, self-service at https://pass.andesprotocol.com).

Discovered
get_tender_detail

Use this before deciding whether a tender is worth pursuing. Pass a tender_id from search_tenders. Returns buyer, description, dates, estimated amount, items, documents, the question deadline and the official source_url, plus agent_guidance: deadlines, questions to ask your user, a checklist of market rules and suggested next calls. If the id is unknown it returns error.code 'not_found': retry with search_tenders using the tender name. Access: open to anonymous agents (rate-limited); a free AndesPass key raises the limit.

Discovered
market_signals

Use this to understand a public procurement market before choosing where to bid: the largest awards, supplier concentration (top-3 share, HHI), fastest-growing categories, award versus estimate, single-bidder and direct-award shares, and live open Compras Ágiles with no quote yet, plus amount_anomalies: awards whose recorded amount is out of range (left out of totals, listed to verify at the source). Each signal has a neutral statement, its number, the change against the previous period, an implication for a supplier and suggested_next_calls. Pass category as a UNSPSC code ('42', '4213') or a name ('equipamiento medico'), and period as last_12m (default), a year ('2025') or a quarter ('2025-Q3'). Built from awarded tenders (OCDS), which lag months: every signal states data_as_of. Anonymous agents get one market-wide signal per call. Access: open to anonymous agents (rate-limited); a free AndesPass key raises the limit.

Discovered
prepare_quote

Use this when your user decided to quote: it builds the kit their person needs to submit, ready for one last click in Mercado Público (Andes cannot submit it). Pass tender_id and the prices: unit_prices per line ([{"line": 1, "unit_price": 12500}], net unless prices_include_vat) or quote_total (VAT included); optionally delivery_days and supplier_id (the user's RUT). Returns the requested lines with the draft (net, IVA 19%, total), readiness checks with their evidence (open, registered supplier, within the buyer's budget, quotes received and their range, delivery, documents to attach), the deadline and the link to submit. It never invents prices: unpriced lines stay unpriced. Any identified agent can call it. If the person you act for approved an AndesPass mandate for procurement.quote.prepare with a ceiling covering the total, the kit says it was prepared under that mandate and the call leaves a signed receipt; without one it is a draft for your person to review. Access: needs an identified agent key (free, self-service at https://pass.andesprotocol.com).

Discovered
search_tenders

Use this first to find ChileCompra public tenders matching a need. Pass Spanish keywords for best results (e.g. 'firewall', 'notebook', 'ciberseguridad'). Returns only tenders still open for bids (set include_closed=true for history), ordered by match score and then the soonest closing date, each with tender_id, retrieved_at and source_url. Andes adds the synonyms Chilean buyers use (e.g. 'ciberseguridad' also finds 'seguridad informática', 'firewall', 'SOC'): audit.expanded_terms lists them and each result's matched_terms says which term hit. One concept per word is best; accents do not matter. audit.index_last_synced_at and audit.index_open_tenders describe the whole index: zero results with a recent sync means nothing open matches today. Call get_tender_detail before deciding anything. Access: open to anonymous agents (rate-limited); a free AndesPass key raises the limit.

Discovered
supplier_history

Use this to size up a supplier, your user's own company or a competitor: what it won, from which buyers, in which categories, how many bidders it beat (competition_when_winning: read median_bidders first, framework agreements with hundreds of bidders skew the average), and, when bidder lists are published, its win rate and most frequent competitors. Pass supplier_id (tax id such as a Chilean RUT, or the supplier_id seen in buyer_history) or supplier_name. Suppliers are grouped by legal identifier; built from awarded tenders in Open Contracting (OCDS) data. An empty matches list comes with a note saying why. Access: needs an identified agent key (free, self-service at https://pass.andesprotocol.com).

Discovered

Recent measurements

Probe history

View status
Latest stored probes. Raw probe rows are kept for 14 days.
CheckedResultResponseTools
Connected1130 ms7